appendix to systemperspective 2035 … power consumption ... 3.3.8 atr - catalytic reformer..... 14...

78
Dok. 17/01970-19 APPENDIX TO SYSTEMPERSPECTIVE 2035 MODELLING OF ENERGY PLANTS

Upload: ngoxuyen

Post on 22-May-2018

212 views

Category:

Documents


0 download

TRANSCRIPT

Dok. 17/01970-19

APPENDIX TO SYSTEMPERSPECTIVE 2035

MODELLING OF ENERGY PLANTS

Dok. 17/01970-19

Indhold

1. Introduction ..................................................................................... 6

2. Heat Areas ....................................................................................... 7

3. Production units .............................................................................. 9

3.1 Dryer Unit ............................................................................................................. 9

3.1.1 Assumptions ............................................................................................. 9

3.1.2 Mass and energy balance: ........................................................................ 9

3.1.3 Power consumption ................................................................................. 9

3.2 Gasification and methanol production ............................................................... 10

3.3 Thermal Gasification and Gas Cleaning Unit ....................................................... 11

3.3.1 Description of processes ........................................................................ 11

3.3.2 Biomass Infeeder .................................................................................... 12

3.3.3 Thermal gasification reactor ................................................................... 12

3.3.4 The Product gas cleaning processes ....................................................... 13

3.3.5 Tar reformer ........................................................................................... 13

3.3.6 Dust filter ................................................................................................ 14

3.3.7 Guard bed (H2S cleaning) ........................................................................ 14

3.3.8 ATR - catalytic reformer.......................................................................... 14

3.3.9 Rectisol CO2 removal .............................................................................. 14

3.4 Methanol Synthesis and Purification .................................................................. 16

3.4.1 Technical description .............................................................................. 16

3.4.2 Recirculation and purge gas ................................................................... 16

3.4.3 Syn gas composition ............................................................................... 16

3.5 Water Shift Reactor Unit ..................................................................................... 18

3.6 Electrolysis Unit .................................................................................................. 19

3.6.1 SOEC ....................................................................................................... 19

3.6.2 Alkaline Electrolysis ................................................................................ 20

Dok. 17/01970-19

3.7 EC gas Splitter Unit ............................................................................................. 21

3.8 GT SC Unit ........................................................................................................... 21

3.9 Steam Turbine Unit ............................................................................................. 22

3.9.1 Cost of Steam Turbine ............................................................................ 23

3.10 Air Separation Unit ............................................................................................. 23

3.11 Gas Boiler Unit .................................................................................................... 24

4. Storages ......................................................................................... 26

4.1 District Heating Storage ...................................................................................... 26

4.2 Oxygen storage ................................................................................................... 26

4.3 SynGas storage ................................................................................................... 28

5. Market Prices ................................................................................ 29

5.1 Oxygen Market ................................................................................................... 29

5.2 Natural gas .......................................................................................................... 29

5.3 Wood Chips ......................................................................................................... 29

5.4 Bio Methanol ...................................................................................................... 29

5.5 Electricity ............................................................................................................ 29

6. Heat demands ............................................................................... 30

6.1 District heating .................................................................................................... 30

6.2 Process Heat ....................................................................................................... 30

6.3 District Heat Sink................................................................................................. 30

7. Simulation Results without ADAPT optimisation .......................... 31

7.1 Energy flows ........................................................................................................ 31

7.2 Power price profile and heat demand per hour ................................................. 32

7.3 Operation profile for the production units ......................................................... 32

7.4 Storage levels ...................................................................................................... 34

8. 34

9. Simulation Results with ADAPT optimisation ............................... 34

Dok. 17/01970-19

9.1 Energy flows ........................................................................................................ 34

9.2 36

9.3 Power price profile and heat demand per hour ................................................. 36

9.4 Operation profile for the production units ......................................................... 36

9.5 Storage levels ...................................................................................................... 38

10. Sensitivity analysis ......................................................................... 39

10.1 Baseline model.................................................................................................... 39

10.1.1 General issues ......................................................................................... 39

10.1.2 Prices ...................................................................................................... 39

10.1.3 CAPEX ..................................................................................................... 39

10.1.4 OPEX ....................................................................................................... 39

10.1.5 Key economic figures.............................................................................. 40

10.1.6 Baseline model Energy Flows ................................................................. 41

10.1.7 41

10.1.8 Baseline model Key Figures .................................................................... 41

10.1.9 42

10.1.10 Baseline model Electrolysis Unit operation ............................................ 42

10.2 43

10.3 43

10.4 Internal Rate of Return (IRR) at different MeOH prices ..................................... 43

10.5 47

10.6 Variation in electricity price and profile ............................................................. 47

10.6.1 More oscillating electricity prices ........................................................... 47

10.6.2 50

10.6.3 Lower electricity prices ........................................................................... 50

10.7 Higher biomass cost ............................................................................................ 54

10.7.1 Biomass price + 20% scenario ................................................................ 54

10.7.2 56

Dok. 17/01970-19

10.7.3 Biomass price + 40% scenario ................................................................ 56

10.8 Forced ASU operation – cost of flexibility ........................................................... 58

10.9 Green Syn Fuel composition on syngas versus baseline composition ................ 61

10.10 Investment cost variations ..................................................................... 64

10.10.1 Investment costs reduced with 30% ...................................................... 64

10.10.2 Investment costs increased with 30% .................................................... 66

10.11 Gas turbine investment halved .............................................................. 68

10.12 ............................................................................................................................ 70

10.13 SOEC versus Alkaline .............................................................................. 70

11. References ..................................................................................... 72

12. Appendix 1 ..................................................................................... 73

12.1.1 Baseline model Energy Flows ................................................................. 73

12.1.2 73

12.1.3 Baseline model Key Figures .................................................................... 73

12.1.4 74

12.1.5 Baseline model Electrolysis Unit operation ............................................ 74

12.1.6 Baseline model Energy Flows new Sifre version ..................................... 75

12.1.7 75

12.1.8 Baseline model Key Figures new Sifre version ........................................ 75

12.1.9 76

12.1.10 Baseline model Electrolysis Unit operation new Sifre version ................ 76

Dok. 17/01970-19

1. Introduction

This document describes modeling of a combined energy plant in 2030/2035.. This type of

energy plant is a part of Energinet’s “System Perspective 2035” analysis. In this analysis two

integrated “Energy Plants” is analysed.

Centralized energy plant (type 3) and Decentralized energy plant (type 2) as shown below.

Figure 1: Integration of power, gas and heat. In this appendix the type 3 plant (upper left) is

documented in more details.

The energy plant is seen as a potential rebuilding/replacement of a central power plant. One of

the constraining parameters for the plant is therefore to supply district heating to the city

where the power plant is located.

The combined energy plant produces: District heating/process heat, methanol for transporta-

tion use and potential peak electricity. Furthermore the energy plant should be able to fit into

an electricity market with large amount of wind energy.

The energy plant will be modeled in Energinet's modeling tool, Sifre. This enable hour to hour

operational optimization based on forecasted hour-to-hour electricity prices.

Sifre’s ADAPT module for investment optimization is used in second step to optimize the sizes

of the individual process units based on one year of operation.

Energy plant type III model:

Dok. 17/01970-19

Figure 2: Sifre detailed model of Energy Plant Type III

This model is partly based on technologies under development. It is assumed, that especially

the Thermal Gasification and Gas Cleaning will be developed to a commercial level with the

today expected investment costs. EA Energianalyse summarise in the report: Grøn Roadmap

2030. Scenarier og virkemidler til omstilling af transportsektorens energiforbrug, November

2015.

”De tekniske udfordringer ved forgasningsteknologierne har vist sig at være betydelige, når

biomasse er råvaren. Et egentligt kommercielt gennembrud kræver derfor en målrettet og lang-

sigtet udviklingsindsats som bedst løftes i internationalt samarbejde. Det synes særdeles usik-

kert, om der kan opnås en betydelig kommerciel produktion inden 2030.” [15]

2. Heat Areas

As Sifre doesn’t work with mass flows and temperature and pressure levels but only energy

flows it has been chosen to operate with three different heat areas:

District Heating area for energy streams at 50-110°C (DH)

Low Temperature Process Heat for energy streams at 110-300°C (LTPH)

High temperature Process Heat for energy streams at 300-1000°C (HTPH)

The general idea of the splitting in three heat areas is, that the HTPH is capable of generating

steam for at steam turbine and supplying HT heat for a SOEC electrolysis. The LTPH is used for

drying and purification processes and can be sold as process heat for industrial use. The DH will

Black numbers are Energy flows in MWh over 24 hours 0

Red numbers are prices in DKK/MWh

0,0

191 0,020

411 0 0

411 To Gas Boiler 0 0

215 To HT Proces Heat

0

5313

4594 5017

0,2866 0,3064 4594 395

0 395

0,0

0

548 To DH

4594

From HTPH 610

0

0,0 503 Ngas 9085

470 0 LTPH 411

12000 12000 9085 4090 10390 10390

0 713

207 207 2903 952 2839 713

2111

2903 988 2839

2338,4 1851

565 1851

1593

From EC

0 539 0,0

380 To Water Shift 258

614,3

1094

230 0

From ST 0,0 From CHP

2596 514,2 519,2 From Water Shift 1372

0

514,16 540

To DH

539 247

610

1081710817 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

be supplying District Heating for the DH network or can be cooled away if in excess. Energy can

freely flow from high temperature areas to lower temperature areas. Thereby the process is

limited by the production of Process Heat, but surplus process heat can be led to the DH sink.

In the following description the DH, LTPH and HTPH will be used for heat streams and Areas.

Dok. 17/01970-19

3. Production units

3.1 Dryer Unit

In the dryer unit, the incoming wood chips, with 50% water content, is dried down to 15% and

grinded. This is necessary to be able to feed the wood into a pressurized gasifier.

Dryer data is found in in the “Polygeneration” report [1], where supplier data for a Metso belt-

drier is chosen.

3.1.1 Assumptions

Wood Chips is supplied at 50% water content and a LHV at: 8.4 MJ/kg

Wood chips are dried down to 15% water content and a LHV at: 15.9 MJ/kg [1]

The process is driven by 2 bar steam, witch in the Sifre model correspond to LTPH (Low Tem-

perature Process Heat). The air and condensate outlet is not implemented in the Sifre simula-

tion.

3.1.2 Mass and energy balance:

Mass Balance:

100 wet wood chips becomes 58.8 dried wood fuel

Energy balance:

100 wet wood chips becomes 111 dried wood fuel

Energy consumption:

Metso bet drier: 3.975 MJ/kg water evaporated [1]

1 MWh wet wood chips = 429 kg wet wood chips. Water to be evaporated is 429*0.412 = 177

kg

Process heat demand: 177*3.975/3600 = 0.195 MWh

3.1.3 Power consumption

In the Polygeneration report [1] a power consumption of 0.01 MWh per 1 MWh wet wood

chips input is used. Same number is used here. NOTE: There are no data on the power con-

sumption for grinding of the dried wood chips. Has to be found when we know the level of

grinding necessary.

The total energy balance of the drying unit will then be:

Dok. 17/01970-19

Figure 3: Energy balance for Dryer Unit

And in Sifre, the fuel input and efficiency will be like shown in figure 4.

Figure 4: Sifre energy balance for Dryer Unit

Sifre input:

Type: Condensation

Production efficiency: B: 3.91

Fuel Consumption: Dried wood: 83%, LTPH: 16.2%, El: 0.83%

ADAPT: Investment cost: 0.36 MDKK/MW [2]

ADAPT: O&M cost: 14,400 DKK/MW/y [6]

ADAPT: Life time: 20 y [G]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 20%/min [G]

Min production: 15% [G]

3.2 Gasification and methanol production

The gasification and methanol production process includes many different production pro-

cesses. In this Sifre model, the processes are grouped in three Production Units:

Thermal Gasification and Gas Cleaning

Water Shift Reactor

Methanol Synthesis and purification

Each unit includes several processes and the input- and output streams for each unit is calcu-

lated as the net input/output for all the processes in the unit. Figure 5 shows the full process

diagram for the biomass to methanol conversion and how the split-up in production units is

chosen.

Dryer Unit

Wet wood: 100

LT Process Heat: 19,5

Electricity: 1

Dried wood: 111

Dryer Unit

Efficiency: 92.1 %

Wet wood: 83.0%

LT Process Heat: 16.2%

Electricity: 0.83%

Dried wood: 92.1%

Dok. 17/01970-19

Figure 5: Process diagram for biomass to methanol process, Methanol via biomass gasification

[3].

3.3 Thermal Gasification and Gas Cleaning Unit

As shown in Figure 5, the Thermal Gasification and Gas Cleaning Unit includes following pro-

cesses:

Biomass infeeder

Thermal gasification reactor

Tar Reformer

Dust filter

Guard bed (H2S cleaning)

ATR - catalytic autothermal reformer

Rectisol CO2 removal

The Water Shift reactor is excluded in this unit because it doesn’t operate with the same load

variations as the rest of the processes. This is due to the Electrolyser Units production of hy-

drogen. This hydrogen makes it possible to partly or fully bypass the Water Shift Reactor espe-

cially at low power prices.

3.3.1 Description of processes

In the following description, the energy in- and outputs are set based on an Aspen+ simulation

model of the process shown. Figures are related to a biomass feed in to the gasification at 100

MW LHV.

Thermal Gasification and Gas Cleaning

Water Shift Reactor

Methanol Synthesis and purification

Dok. 17/01970-19

3.3.2 Biomass Infeeder

Type: Piston infeeder.

Key figures [1]

P: 440 kW

3.3.3 Thermal gasification reactor

Type: The biomass is gasified in pressurized conditions (at 25 bar) and in presence of oxygen

and steam. Pressurization of the whole gas production process through a pressurized biomass

feeding system introduces significant savings in the subsequent gas compression required to

achieve the optimal synthesis pressures both in terms of capital and operating costs. However,

since combustion and gasification occur in the same reactor, to avoid large amount of inert

nitrogen the direct gasification concepts necessitates a pure oxygen stream. Gasification oc-

curs also in presence of steam which is required as a reforming agent. These latter endother-

mic reactions require energy to be provided by combustion and therefore the consumption of

oxygen increases with the steam input. The optimal ratio between oxygen and steam for a

temperature around 886°C and for a pressure of around 25 bar is around 1:1 [4].

Mass balance of the TG process:

Both Process Steam and oxygen is supplied at 25 bar pressure and around 230°C. In this tem-

perature area LTPH can be utilized.

3.3.3.1 Energy Balance

The energy balance is the one that controls the simulations in Sifre. Therefore every significant

stream must be assigned with energy. O2 does have a LHV at 0 MJ/kg. If this is used it will not

be possible to track the O2 production and consumption. The O2 stream from Electrolyser to

Gasification is vital and to be able to track the O2 production and consumption, it is necessary

to assign the O2 an LHV. It is chosen to set an arbitrary LHV value for O2 to: 0.001 MJ/kg. Low

enough not to corrupt the general energy balance but high enough to be calculated correctly.

In Sifre it has to be assigned as a percentage of energy input. 1 MJ of wood (0.063 kg) will re-

quire 0.0184 kg of O2. As the LHV of O2, is set to 0.001 MJ/kg, 1 MJ of wood will require

0.0000184 MJ of O2 in Sifre.

Now the energy balance for Sifre can be concluded, taking into account 0.33 MW LTPH for

Oxygen preheating (Not with SOEC):

Thermal

Gasification

Dried wood: 22.27

LT Process Steam: 6,57

Oxygen: 6,57

Product gas: 35.40

Figure 6: Mass balance for Thermal Gasification [3]

Dok. 17/01970-19

Figure 7: Energy balance for Thermal Gasification

The estimated composition of the product gas is shown in Figure 8:

Figure 8: Estimated composition of product gas [3]

3.3.4 The Product gas cleaning processes

The mass and energy balances for the individual steps of the product gas cleaning process will

not be outlined fully here. Only the sum-up of the processes will be presented. But as the

technology choices for these steps are important for the energy balance the, the technology is

briefly described and key figures presented.

3.3.5 Tar reformer

The Tar reformer is an isothermal (890°C) catalytic bubbling fluidized bed in which the tar

compounds are reformed in presence of the abundant steam content of the product gas. Data

about such a reformer were obtained from publications by the US NREL (Spath, Aden et al.

2005) [5]. Tar compounds such as Naphthalene, light hydrocarbons such as propane and

ethane as well as ethylene and acetylene are found in the product gas from the gasifier and are

largely reformed into H2 and CO by catalytic cracking. Methane is also partially cracked alt-

Thermal

Gasification

Dried wood: 100

LT Process Heat: 5.45

Oxygen: 0.00184

Product gas: 105

Dok. 17/01970-19

hough about half of that still remains in the gas at the reformer outlet. The heat for reforming

is provided by circulating the bed and the catalyst from a side combustor which is fuelled by a

certain quantity of product gas (about 10% of the total product gas) that is by-passed prior the

reformer and therefore does not contribute to methanol production [3].

Key figures [3]:

Energy efficiency (product gas to syngas): 90%

Heat recovery: 4.6 MW HTPH

Power consumption: 1.74 MW (for air compressor)

3.3.6 Dust filter

At the outlet of the tar reformer, particulate matter, alkali materials, and sulphur compounds

are still present in the gas and must be removed prior to the gas upgrading and synthesis reac-

tions. This is done by hot gas cleaning technologies and in particular by candle filters. The syn

gas has to be cooled prior to the filter

Key figures [3]:

Energy efficiency: 100%

Heat recovery: 9.92 MW HTPH (cooling prior to filter)

3.3.7 Guard bed (H2S cleaning)

The sulphur, assumed here completely in the form of H2S, is removed through a guard bed

based on metal oxides.

3.3.8 ATR - catalytic reformer

In order to finally convert the remaining hydrocarbons that would otherwise remain as inert in

the methanol synthesis process, a catalytic reformer is used. Steam injection is not required as

the steam to carbon ratio is already higher than 1. Oxygen is added for the cracking. The re-

former operates at temperatures higher than the dust filter, so syn gas heating is required.

After the reforming the temperature has to be lowered before the Water Shift Reactor.

Key figures [3]:

Oxygen addition: 1.98 t/h

Oxygen heating (not with SOEC electrolysis): 0.09 MW LTPH

Syn gas heating: 6.62 MW HTPH

Syn gas cooling: 11.56 MW HTPH

3.3.9 Rectisol CO2 removal

The gas is cooled to ambient temperature and most of the CO2 is removed by a Rectisol pro-

cess where methanol is used as a physical absorbent. A final CO2 concentration of 3% in the

dry syngas is obtained as it is the optimal concentration for subsequent methanol synthesis [3].

Dok. 17/01970-19

Key figures [3]:

Syn gas cooling: 10.12 MW LTPH

Heat added to process: 0.2 MW LTPH

Power: 0.76 MW

This cooling is correct if no gas passes the Water Shift Reactor. The extra cooling needed, if

some gas passes the Water Shift Reactor is allocated to the Water Shift Reactor Unit.

Adding all the gas cleaning process steps up results in this energy balance:

Figure 9: Sum-up energy balance for gas cleaning

In total the Thermal Gasification and Gas Cleaning Unit looks like this:

Figure 10: Total energy balance for Thermal Gasification and gas Cleaning.

As Sifre only can handle two output streams from a Production Unit, is has been necessary to

group the heat outputs in HeatMix1 and split them in a subsequent Heat Splitter1.

Sifre input (Thermal Gasification and Gas Cleaning Unit):

Type: Backpressure (Syn gas and HeatMix1)

Cb: 3.129

Production efficiency: B: 4.895

Fuel Consumption: Dried wood: 97.14168%, Oxygen: 0.00232%, El: 2.856%

ADAPT: Investment cost: 4.70 MDKK/MW [2]

ADAPT: O&M cost: 188,000 DKK/MW/y [6]

ADAPT: Life time: 20 y [2]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost: 17.9 DKK/MWh [8] (only for gasification unit)(modeled in Sifre as a

tax)

Ramping up/down: 50%/min [6]

Min production: 15% [6][G]

Emissions: Has to be estimated

Product gas

cleaning

Product gas: 105

Power: 2.5 HTPH: 19.48

Product gas: 75.7

LTPH: 9.83

Thermal Gasi-

fication and

Gas Cleaning

Dried wood: 100

Power: 2.94 HTPH: 19.48

Product gas: 75.7

LTPH: 4.71 Oxygen: 0.00239

Oxygen: 0.00055

Dok. 17/01970-19

Sifre input (HeatSplitter1):

Type: Backpressure (HTPH and LTPH)

Cb: 4.136

Production efficiency: B: 4.47

Fuel Consumption: HeatMix1: 100%

All other inputs are 0, as this component isn’t a physical component

3.4 Methanol Synthesis and Purification

The total process of methanol synthesis and purification is shown in Figure 5. All processes

included will be handled as one Production Unit in Sifre. This is a fair simplification as all the

processes energy consumptions and performances are directly related to the inflow of syngas

and its composition. The composition is kept fixed by adjusting the Hydrogen inflow to an op-

timal mixture in all operation points.

3.4.1 Technical description

Methanol synthesis occurs in a fixed bed reactor at 90 bar [3] where steam is produced by

reactor cooling. The reaction products are subsequently cooled leaving a two-phase stream.

The crude methanol is obtained at the bottom of the flash drum and is then sent to distillation

after the pressure has been released to atmospheric values. A stripper is used to evaporate the

remaining gases entrained in the raw methanol. The methanol rich liquid still present a large

quantity of water which is removed in a two column system.

3.4.2 Recirculation and purge gas

A certain amount of hydrogen is recovered from a purge stream after the methanol synthesis

which is mixed to the fresh syngas feed to reach the final synthesis ratio of around 2.05 with a

3% CO2 molar fraction [3]. The syngas is then mixed with the synthesis loop gas obtained at the

top of the flash drum right after the reactor. Here the synthesis loop is assumed to be 3:1 in

volume with respect to the fresh syngas. The purge gases are burnt thus producing additional

heat that can be used for steam production.

3.4.3 Syn gas composition

The methanol synthesis operates with highest efficiency, if the M ratio is 2.05 [3]. Further-

more, the CO2/CO ratio should be low [2]. The M ratio is calculated as shown in Figure 11:

Figure 11: M-ratio calculation by mole fraction

The H2/CO ratio of the raw syngas is: 20.1/15.8 = 1.27

Though the H2 level has to be raised. As described some hydrogen is recirculated from the

purge gas. This recirculation raises the M-ratio 0.13. So the M- ratio before the hydrogen is

added has to be: 1.92.

As the CO2 content is 3% after the CO2 scrubber, the H2 content can be calculated:

Dok. 17/01970-19

1.92 = (x-3)/(97-x+3)

x = 66.8

The mole fraction of hydrogen has to be 66.8% at the inlet to the methanol synthesis and puri-

fication unit. Without Water Shift Reaction the mole fraction of H2 is: 51.2% and the mole frac-

tion of CO is: 45.9%. To raise the level to 66.8% further 45.6 mole of H2 has to be added per

100 mole raw syn gas.

Approximately one Mole of H2 has to be added for each mole CO in the syngas.

This has to be converted to energy terms to be used in Sifre. As CO represents approximately

46% of the syngas after CO2 and H2O removal, approximately 0.46 Mole of H2 has to be added

to 1 Mole of cleaned syngas.

LHV:

CO: 283 kJ/Mole, H2: 244 kJ/Mole

1 Mole of cleaned syn gas has a LHV at: 255 kJ/Mole

0.46 Mole of H2 has a LHV at: 112 KJ

Thus the energy input to Methanol synthesis unit has to be: 69% cleaned syn gas and 31 % H2.

By fixing this ratio in the Production Unit for Methanol Synthesis in Sifre, the simulation forces

to model to produce the needed H2 either by Electrolysis or by Water Shift Reaction.

The power consumption is quit high due to syn gas compression (90 bar) prior to methanol

synthesis.

Added up power consumption: 2.17 MW [3]

There are a lot of processes where heat either has to be added or removed from the stream.

Summed up following heat demand/production is achieved:

Net heat output [3]:

HTPH: 1.48 MW

LTPH: 9.16 MW

DH: 5.68 MW

This gives a net energy balance for the Methanol Synthesis and Purification. The Energy bal-

ance in Figure 12 is based on a set up with no electrolysis hydrogen.

Figure 12: Net energy balance for Methanol Synthesis and Purification (no H2 from Electrolysis).

Methanol

Synthesis and

Purification

Raw Syn Gas: 52.2

H2 from WSH: 23.5 HTPH: 1.48

Methanol: 59.7

LTPH: 9.16 Power: 2.17

DH: 5.68

Dok. 17/01970-19

As Sifre only can handle two output streams from a Production Unit, is has been necessary to

group the heat outputs (HeatMix2 and 3) and split them in two subsequent Heat Splitters (2

and 3).

Sifre input (Methanol Synthesis and Purification Unit):

Type: Backpressure (Methanol and HeatMix2)

Cb: 3.66

Production efficiency: B: 4.696

Fuel Consumption: Syn gas: 67%, H2: 30.18%, El: 2.787%

ADAPT: Investment cost: 1.32 MDKK/MW [2]

ADAPT: O&M cost: 39,600 DKK/MW/y [6]

ADAPT: Life time: 20 y [2]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 25%/min [G]

Min production: 15% [G]

Sifre input (HeatSplitter2):

Type: Backpressure (HeatMix3 and DH)

Cb: 1.873

Production efficiency: B: 5.522

Fuel Consumption: HeatMix2: 100%

All other inputs are 0, as this component isn’t a physical component

Sifre input (HeatSplitter3):

Type: Backpressure (HTPH and LTPH)

Cb: 0.162

Production efficiency: B: 325.82

Fuel Consumption: HeatMix3: 100%

All other inputs are 0, as this component isn’t a physical component

3.5 Water Shift Reactor Unit

As mentioned under the Methanol Synthesis and Purification description the methanol produc-

tion needs more H2 the available in the raw syn gas. The H2 can come from electrolysis but in

periods with high power prices is could be more feasible to produce the extra H2 from conver-

sion of CO in the raw syn gas to H2. This can be done in a Water Shift Reaction where the fol-

lowing process occurs:

CO + H2O -> H2 + CO2

The H2O is added as high pressure high temperature steam (HTPH). The process is exothermic

so cooling down of the produced gas brings energy to the LTPH area. Based on data from [3]

the energy balance of the water shift reaction becomes:

Dok. 17/01970-19

Figure 13: Energy balance Water Shift Reactor Unit

Sifre input:

Type: Backpressure (Hydrogen and LTPH)

Cb: 2.745

Production efficiency: B: 5.96

Fuel Consumption: SynGas: 82.37%, HTPH: 17.63%

ADAPT: Investment cost: 0.54 MDKK/MW [2]

ADAPT: O&M cost: 26,200 DKK/MW/y [6]

ADAPT: Life time: 20 y [2]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 25%/min [G]

Min production: 15% [G]

3.6 Electrolysis Unit

The electrolysis unit converts electricity (and process heat) to Hydrogen and Oxygen (and pro-

cess heat).

Two electrolysis technologies will be simulated:

3.6.1 SOEC

A high temperature concept that have high electricity to Hydrogen ratio. The technology is

under development. The energy balance for the SOEC technology is taken from The Technolo-

gy Data Catalogue for Energy Plants-aug. 2016 [7]:

Figure 14: Energy balance for SOEC electrolysis

The temperature of the heat source should be the same as the working tempera-

ture, i.e. 800 – 1000°C [7].

Water Shift

Reactor

Syn gas: 100

HTPH: 21.4

H2 rich gas: 73.3

LTPH: 26.7

Dok. 17/01970-19

3.6.2 Alkaline Electrolysis

This is the market standard today and has been commercial available for almost hundred year.

The energy balance for the SOEC technology is:

Figure 15: Energy balance for Alkaline Electrolysis [7]

Sifre can only handle two output streams. Therefore the Electrolysis is modeled with two Pro-

duction Units. One unit producing heat and EC gas (mixture of H2 and O2) and an EC gas Splitter

converting the EC gas to H2 and O2.

Sifre input (SOEC):

Type: Condensation (EC gas)

Production efficiency: B: 4.41

Fuel Consumption: El (max): 100 %, HTPH (max): 16.67% [7]

ADAPT: Investment cost: 4.4 MDKK/MW [7]

ADAPT: O&M cost: 109,888 DKK/MW/y [7]

ADAPT: Life time: 20 y [7]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 100%/min [G]

Min production: 15% [G]

Sifre input (Alkaline):

Type: Backpressure (ECGas and DH)

Cb: 3.87 [7]

Production efficiency: B: 6.207 [7]

Fuel Consumption: el: 100%

ADAPT: Investment cost: 7.4 MDKK/MW [7]

ADAPT: O&M cost: 296,000 DKK/MW/y [7]

ADAPT: Life time: 30 y [7]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 100%/min [G]

Min production: 15% [G]

Dok. 17/01970-19

3.7 EC gas Splitter Unit

The EC gas Splitter is set to Back pressure Plant to fix the relation between H2 and O2.

H2 set to primary production.

Cb is difficult to set. It ought to be infinity as all the energy follows the H2 stream. But to be

able to track the O2 production and consumption, it is necessary to assign the O2 an LHV. It is

chosen to set an arbitrary LHV value for O2 to: 0.001 MJ/kg. Low enough not to corrupt the

general energy balance but high enough to be calculated correctly (I hope)

Mass balance for EC gas:

1000 EC gas = 111 H2 + 889 O2.

LHV for H2 is 120 MJ/kg. LHV for O2 is chosen to: 0,001 MJ/kg.

Energy balance:

1000 EC gas = 999.933 H2 + 0.0667 O2

That results in a cb-value of: 999.933/0.0667 = 14991

The efficiency is set to 99.9933%

Sifre input:

Type: Backpressure (Hydrogen and Oxygen)

Cb: 14991

Production efficiency: B: 3.60024

Fuel Consumption: EC gas: 100%

All other inputs are 0, as this component isn’t a physical component

3.8 GT SC Unit

It is chosen to use a simple cycle GT unit for peak load power production. The advances of the

simple cycle GT is fast regulation (0-100% in 15 min. typically) [8] and low investment cost per

MWe.

The electric efficiency of a large (40-125 MWe) single cycle GT is in the Technology Data Cata-

logue for Energy Plants- UPDATE 2016 [8] estimated to 41% in 2030 on an annual average and

Cb = 1. The exhaust flue gas can be directed to the HTPH area for use in the Methanol process

or in the Steam Turbine if capacity is available. Utilizing the flue gas heat only in the HTPH area

expects to raise the Cb to 1.5. The link from the HTPH to DH and heat sink makes it possible to

ramp up the GT faster than the Steam Turbine and thereby keep the regulation capacity.

Sifre input [8]:

Type: Backpressure (El and HTPH)

Cb: 1.5 [G]

Production efficiency: B: 8.78

Fuel Consumption: Syn Gas: 100%, NGas: 100%

ADAPT: Investment cost: 4.17 MDKK/MW

Dok. 17/01970-19

ADAPT: O&M cost: 138,570 DKK/MW/y

ADAPT: Life time: 25 y

Maintenance: 2.5 Weeks/y

Outage Probability: 2%

Operating Cost: 31.3 DKK/MWh (modeled in Sifre as a tax)

Ramping up/down: 20%/min

Min production: 15%

Emissions: NOx: 10g/GJ, CH4: 1.5 g/GJ, N2O: 1 g/GJ

3.9 Steam Turbine Unit

The biomass to methanol process generates a high amount of high temperature heat, especial-

ly from syn gas cooling. These heat outputs are gathered in the HTPH Area. High temperature

heat can produce high pressure and high temperature steam for power production. It has been

chosen to implement a Steam Turbine unit for utilizing the excess HTPH. It is assumed, that all

excess HTPH can be converted to 100 bar steam at 500°C [3]. This is let to the Steam Turbine

Unit that is set up as an extraction turbine for DH production if needed.

Shows a pinch diagram for a quit similar process [3]:

Figure 16: Pinch diagram for similar biomass to methanol proces [3]

Figure 17: Input / outputs for Steam Turbine Unit

Steam Tur-

bine HTPH

Electricity

DH

Dok. 17/01970-19

The split between electricity and DH is flexible. The Unit is modeled with a max electricity effi-

ciency at 45% and Cv = 0.15 and Cb = 1 [8].

3.9.1 Cost of Steam Turbine

The investment cost of steam generation and steam turbine is taken from the NREL report:

Gasoline from Wood via Integrated Gasification, Synthesis, and Methanol-to- Gasoline Technol-

ogies, Steven D. Phillips et al., 2011. [9]

In this study the total installed cost of steam generation and extraction type steam turbine is

M23.1 USD.

From the mass- and energy balance is the power generation found to 35 MWe at a steam

pressure at 33 bar and a steam temperature at 482°C. In this Sifre model the steam turbine

produce approximately 50 MW and the steam conditions are: 500°C and 50 bar. These sizes

and conditions are in the same range and therefore, is it assumed, that the specific prices of

investment in the US study can be used. The investment cost is in 2007 USD. In 2007 the price

in EUR is M16.9 EUR1. The specific price then becomes: 0.48 MEUR/MWe

Sifre input:

Type: Extraction (El and DH)

Cb: 1, Cv: 0.15 [8]

Production efficiency: B: 8 [8]

Fuel Consumption: HTPH: 100%

ADAPT: Investment cost: 3.59 MDKK/MWe (2007) [9]

ADAPT: O&M cost: 107,700 DKK/MW/y [6]

ADAPT: Life time: 30 y [G]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 5%/min [G]

Min production: 15% [G]

3.10 Air Separation Unit

The air separation unit is producing high pressure oxygen from air. The only input to the pro-

cess is electricity. Based on the paper: Oxygen Specific Power Consumption Comparison for Air

Separation Units by Yas A. Alsultanny and Nayef N. Al-Shammari [10] the specific power con-

sumption for an ASU type 31 is 0.608 KWh/Nm3 O2. This has to be recalculated to energy

streams, and with an arbitrary LHV for O2 set to 0.001 MJ/kg and a weight of O2 at 1.429

kg/Nm3, the results becomes: 5508 GJ/MWh O2. This is a very low “energy efficiency” but it

does not give any sense to talk about efficiencies as the LHV of O2 is set arbitrary.

1 http://www.x-rates.com/average/?from=EUR&to=USD&amount=1&year=2007: 1.37

USD/EUR

Dok. 17/01970-19

The investment cost of the ASU is based on data from the NREL report: Techno-Economic Anal-

ysis of Biofuels Production Based on Gasification, Ryan M. Swanson et al., 2010 [11].

The ASU in this report has a production rate at 714 t/d O2 and costs 24.3 MUSD. With a LHV for

O2 set to 0.001 MJ/kg the specific investment cost will be: 2942 MUSD/MW = 15988

MDKK/MW

Sifre input (Option 1):

Type: Condensation (Oxygen)

Efficiency: B: 5508

Fuel: Electricity

ADAPT: Investment cost: 15988 MDKK/MW (2007) [11]

ADAPT: O&M cost: 480,000,000 DKK/MW/y [6]

ADAPT: Life time: 20 y [G]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 5%/min [G]

Min production: 15% [G]

Another ASU option is to generate liquid oxygen (ASU-51) [10]. This option has higher energy

consumption per t O2 but the liquid O2 makes it less costly to store. The cost of such a plant is

approximately the same as option 1, but the efficiency is changed. The energy consumption for

an ASU-51 is 0.772 KWh/Nm3 O2 [10].

Sifre input (Option 2):

Type: Condensation (Oxygen)

Efficiency: B: 7001

Fuel: Electricity

ADAPT: Investment cost: 15988 MDKK/MW (2007) [11]

ADAPT: O&M cost: 480,000,000 DKK/MW/y [6]

ADAPT: Life time: 20 y [G]

Maintenance: 2 Weeks/y [G]

Outage Probability: 2% [G]

Operating Cost:

Ramping up/down: 5%/min [G]

Min production: 15% [G]

3.11 Gas Boiler Unit

The gas boiler unit produces LTPH from three different gas sources. Ngas, SynGas and H2. It is

assumed, that the boiler can operate with all possible mixtures of the three gas sources.

Sifre input [8]:

Type: Heatboiler

Efficiency: B: 3.46

Dok. 17/01970-19

Fuel: Ngas, SynGas, H2

ADAPT: Investment cost: 0.37 MDKK/MW

ADAPT: O&M cost: 14,155 DKK/MW/y

ADAPT: Life time: 25 y

Maintenance: 0.4 Weeks/y

Outage Probability: 1%

Operating Cost: 7.45 DKK/MWh (modeled in Sifre as a tax)

Ramping up/down: 15%/min

Min production: 15%

Dok. 17/01970-19

4. Storages

4.1 District Heating Storage

The energy content of a typical district heating storage: 70 kWh/m3 [8]

Investment: 1192 DKK/m3 [8]

Sifre input [8]:

Type: DH storage

ADAPT: Investment cost: 0.017 MDKK/MWh

ADAPT: O&M cost: 0 DKK/MW/y

ADAPT: Life time: 30 y [G]

Maintenance: 0 Weeks/y

Outage Probability: 0%

Operating Cost:

Charge rate: 200 MW [G]

Discharge rate: 200 MW [G]

Start-up capacity : 10 MWh (to avoid very high DH price in the first hour)

Charge efficiency: 100%

Discharge efficiency: 100%

Loss: 0.0001 %/h

4.2 Oxygen storage

It has been very difficult to find investment costs for pressurerized oxygen storage vessels. The

best available data, is has been able to find under the given timeframe is data for compressed

air storage. It is assumed, that a storage vessel for compressed air also can be used for Oxygen.

The found vessel is a 3105 m3 vessel operating at 103 bar. This is a fine match as many ASU

unit deliver the oxygen at 90 bar (has to be checked, if the energy consumption for the used

ASU concept includes compression to 90 bar).

The purchase cost for this vessel is estimated to M16 USD and the installed cost to M49 USD

[12]. As this storage “vessel” consists of 34 storage vessels, it is assumed, that the pricing for

other sizes are linear.

To calculate the amount of oxygen stored in such a vessel, The Ideal Gas Law is used:

PV=nRT, R= 0.082, P: pressure in atm, V: volume in liter, n: number of mole gas, T: temperature

in Kelvin.

n = PV/RT

n = 88.8*3105000/0.082*310 = 10,846,735 mole O2

To be able to deliver gas to the pressurised gasifier, it is assumed, that the lowest pressure in

the O2 storage is 30 bar. The content of the storage is then:

n = 29.6*3105000/0.082*310 = 3,615,578 mole O2

The active storage capacity is therefore: 7.23 mill mole O2 that equals 231.4 t O2

Dok. 17/01970-19

With the arbitrary LHV on O2 set to 0.001 MJ/kg, the “energy” content of the storage is: 0.0643

MWh

The specific cost of the storage then becomes: 274 MDKK/0.0643 MWh = 4261 MDKK/MWh

Sifre input (option 1):

Type: O2 storage

ADAPT: Investment cost: 4261 MDKK/MWh [12]

ADAPT: O&M cost: 0 DKK/MW/y [G]

ADAPT: Life time: 30 y [G]

Maintenance: 0 Weeks/y

Outage Probability: 0%

Operating Cost:

Charge rate: 1 MW

Discharge rate: 1 MW

Charge efficiency: 100%

Discharge efficiency: 100%

Loss: 0 %/h

If liquid oxygen is produced for storage (ASU-51) the storage is a totally different kind. The

storage then should not be able to manage high pressure but instead very low temperature, as

Oxygen is liquid below -183°C. The density is much higher than compressed Oxygen. The stor-

age tank in option 1 can contain 231 t O2 while the same storage volume can contain 3,543 ton

liquid oxygen. It has not been possible to find investment cost for liquid O2 storage in this pro-

jects timeframe, but it is recommended to get such budget data from tank suppliers. For now

the tank cost for compressed oxygen/air storage is used. But as the density is 15 times higher

the cost per MWH is estimated to be 15 times lower. Normally there is a need for heat supply

for evaporation before use in gasification but it is assumed that mixing with the main stream

oxygen from Electrolysis at 850-1000°C will supply enough heat.

Sifre input (option 2):

Type: O2 storage

ADAPT: Investment cost: 284 MDKK/MWh [12]

ADAPT: O&M cost: 0 DKK/MW/y [G]

ADAPT: Life time: 30 y [G]

Maintenance: 0 Weeks/y

Outage Probability: 0%

Operating Cost:

Charge rate: 1 MW

Discharge rate: 1 MW

Charge efficiency: 100%

Discharge efficiency: 100%

Loss: 0 %/h

Dok. 17/01970-19

4.3 SynGas storage

It has been very difficult to find investment costs for pressurerized gas storage vessels. The

best available data, is has been able to find under the given timeframe is data for compressed

air storage. It is assumed, that a storage vessel for compressed air also can be used for com-

pressed SynGas. The found vessel is a 3105 m3 vessel operating at 103 bar. The SynGas is pro-

duced at 25 bar. In the Methanol Synthesis the syngas is compressed up to 90 bar. It is as-

sumed, that the storage at this point in the process and therefore a storage at 90 bar will re-

quire no extra compression of the syngas at inlet to storage. A minor compression at the outlet

of the storage is necessary to 90 bar again, if the storage level is low. The energy consumption

for this compression is not jet included in the simulation. The temperature just before the

Methanol Synthesis is approximately 60°C [3]

The purchase cost for this vessel is estimated to M16 USD and the installed cost to M49 USD

[12]. As this storage “vessel” consists of 34 storage vessels, it is assumed, that the pricing for

other sizes are linear.

To calculate the amount of SynGas stored in such a vessel, The Ideal Gas Law is used:

P*V=n*R*T, R= 0.082, P: pressure in atm, V: volume in liter, n: number of mole gas, T: temper-

ature in Kelvin.

n = P*V/R*T

n = 88.8*3105000/0.082*333 = 10,097,561 mole SynGas

To reduce the power consumption for compression at the outlet it is assumed, that the storage

operates between 90 bar and 30 bar. The content of the storage is then:

n = 29.6*3105000/0.082*333 = 3,365,854 mole SynGas

The active storage capacity is therefore: 6.73 mill mole SynGas that equals 75.7 t SynGas (11.25

g/Mole [3])

With LHV on Syngas just before Methanol Synthesis at MJ/mole, the energy content of the

storage is: 457.7 MWh.

The specific cost of the storage then becomes: 274 MDKK/457.7 MWh = 0.60 MDKK/MWh

Sifre input:

Type: SynGas storage

ADAPT: Investment cost: 0.6 MDKK/MWh [12]

ADAPT: O&M cost: 0 DKK/MW/y [G]

ADAPT: Life time: 30 y [G]

Maintenance: 0 Weeks/y

Outage Probability: 0%

Operating Cost:

Charge rate: 100 MW

Discharge rate: 100 MW

Charge efficiency: 100%

Discharge efficiency: 100%

Loss: 0 %/h

Dok. 17/01970-19

5. Market Prices

5.1 Oxygen Market

An Oxygen Market has been added to receive excess O2. The oxygen market could be just flar-

ing the oxygen or it could be selling it. Price is set to 0 DKK/MWh

5.2 Natural gas

Ngas prices are assumed to be constant over the evaluated period.

Ngas: 59.7 DKK/GJ in 2030 in 2016 prices [14]

5.3 Wood Chips

WoodChip prices are assumed to be constant over the evaluated period.

Wood Chips: 57.5 DKK/GJ in 2030 in 2016 prices [14]

5.4 Bio Methanol

Bio Methanol prices are assumed to be constant over the evaluated period.

EA-analyse has estimated the future production cost of 2. Generation bioethanol to be 230

DKK/GJ (828 DKK/MWh) [14]. This price is assumed to be the upper limit. The lower limit is the

expected gasoline price. ENS estimates this price to be 129.1 DKK/GJ in 2030 and the associat-

ed CO2 cost to be 37 DKK/GJ gasoline. The lower limit is therefore set to: 166.1 DKK/GJ (598

DKK/MWh)

In this study average of the two limits has been chosen as the basic price: 198.1 DKK/GJ (713

DKK/MWh)

5.5 Electricity

The electricity price is assumed to follow the price profile in Energinet's Analyseforudsætninger

2016: AF2016 + 50 DKK/MWh as a tax. The tax is put on the electricity price because the total

plant is a net consumer of electricity.

Dok. 17/01970-19

6. Heat demands

6.1 District heating

Profile: Varmeprofil

Type: non Flexible

Amount: 500,000 MWh/year

6.2 Process Heat

Profile: Varmeprofil

Type: non Flexible

Amount: 100,000 MWh/year

6.3 District Heat Sink

The district heat sink is put in the model to be able to operate the biomass to methanol pro-

cess also when there is very low heat demand. The heat sink could be seen as a “summer cool-

er”. The demand is set very high and the type is set to price cutting to make the District Heat

Sink taking excess heat, when the price comes near zero.

Profile: Varmeprofil

Type: Price cutting (0.1 DKK/MWh)

Amount: 10,000,000 MWh/year

NB: Sifre calculated in general very high costs of the heat streams and in some hours extremely

high prices. It has not been possible in this project to figure out why. Because of that the prices

for the heat streams are fixed in the economy calculation. The high prices in the internal heat

streams do not affect the optimization of the operation or equipment sizes [Thomas Sejr Jen-

sen]

Dok. 17/01970-19

7. Simulation Results without ADAPT optimisation

The described energy system is simulated in Sifre. To show how Sifre optimize the operation

with regard to the electricity price, the operation of all major Production Units and energy

storages are shown for one day. The day chosen is the 4.th of January 2030.

7.1 Energy flows

In Figure 18 the total amount of energy transported between the different Production Units

and Areas is shown.

Figure 18: Simulated process without ADAPT optimization

In the following graphs the Production Units and Energy Storages operation variations is

shown.

Black numbers are Energy flows in MWh over 24 hours 0

Red numbers are prices in DKK/MWh

0,0

191 0,020

411 0 0

411 To Gas Boiler 0 0

215 To HT Proces Heat

0

5313

4594 5017

0,2866 0,3064 4594 395

0 395

0,0

0

548 To DH

4594

From HTPH 610

0

0,0 503 Ngas 9085

470 0 LTPH 411

12000 12000 9085 4090 10390 10390

0 713

207 207 2903 952 2839 713

2111

2903 988 2839

2338,4 1851

565 1851

1593

From EC

0 539 0,0

380 To Water Shift 258

614,3

1094

230 0

From ST 0,0 From CHP

2596 514,2 519,2 From Water Shift 1372

0

514,16 540

To DH

539 247

610

1081710817 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

7.2 Power price profile and heat demand per hour

Figure 19: Electricity price 4th Jan. 2030

Figure 20: Heat Demand 4th of Jan. 2030

7.3 Operation profile for the production units

Figure 21: GT operation 4th of Jan. 2030

Figure 22: Electrolysis operation 4th of Jan

2030

Figure 23: Gasification operation 4th of Jan

2030

Figure 24: Methanol Synthesis operation 4th

of Jan 2030

Dok. 17/01970-19

Figure 25: ST operation 4th of Jan 2030

Figure 26: Gasboiler operation 4th of Jan 2030

Figure 27: WSR operation 4th of Jan 2030

Figure 28: ASU operation 4th of Jan 2030

Dok. 17/01970-19

7.4 Storage levels

Figure 29: DH storage level operation 4th of

Jan 2030

Figure 30: Syn Gas storage level operation 4th

of Jan 2030

Figure 31: O2 storage level operation 4th of

Jan 2030

8. Simulation Results with ADAPT optimisation

The described energy system is simulated in Sifre utilizing the ADAPT function to optimize the

size of each Production Unit and each Energy Storage. To lower the amount of data only one

day is simulated. It is the 4.st of January 2030. The investment optimization is performed for

only one day of operation. This is not the way to determine the best sizes for the processes,

but only to make the two simulations with and without ADAFT comparable.

8.1 Energy flows

In Figure 32 the total amount of energy transported between the different Production Units

and Areas is shown.

Dok. 17/01970-19

Figure 32: Simulated process with ADAPT optimization

In the following graphs the Production Units and Energy Storages operation variations is

shown.

Black numbers are Energy flows in MWh over 24 hours 0

Red numbers are prices in DKK/MWh

0,0

215 0,000

7392 7392 3031

0 To Gas Boiler 0 0

215 To HT Proces Heat

2020,5

1028

4091 4350

0,2729 0,2729 4091 395

0 395

0,0

0

1038 To DH

4090

From HTPH 661

0

0,0 0 Ngas 8649,4

294 0 LTPH 0

11425 11425 8649 3894 9892 9891,7

0 713

207 207 2764 0 2703 713

2009

2764 941 2703

2226,3 1762

538 1762

1516

From EC

0 6493 0,0

306 To Water Shift 246

1061,7

588

306 2020,5

From ST 0,0 From CHP

2596 1067 519,2 From Water Shift 2769

0

1067,5 1086

To DH

6493 310

753

1029910299 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

8.2 Power price profile and heat demand per hour

Figure 33: Electricity price 4th Jan. 2030

Figure 34: Heat Demand 4th of Jan. 2030

8.3 Operation profile for the production units

Figure 35: GT operation 4th of Jan. 2030

Figure 36: Electrolysis operation 4th of Jan

2030

Figure 37: Gasification operation 4th of Jan

2030

Figure 38: Methanol Synthesis operation 4th

of Jan 2030

Dok. 17/01970-19

Figure 39: ST operation 4th of Jan 2030

Figure 40: Gasboiler operation 4th of Jan

2030

Figure 41: WSR operation 4th of Jan 2030

Figure 42: ASU operation 4th of Jan 2030

Dok. 17/01970-19

8.4 Storage levels

Figure 43: DH storage level operation 4th of

Jan 2030

Figure 44: Syn Gas storage level operation 4th

of Jan 2030

Figure 45: O2 storage level operation 4th of

Jan 2030

Dok. 17/01970-19

9. Sensitivity analysis

9.1 Baseline model

The baseline model for the Energy Plant Type III simulation is briefly described and the energy

flows and standard output data are presented.

9.1.1 General issues

The baseline model includes all processes described in this document. For the ASU, option 2 is

chosen with production of liquid oxygen and therefore also option 2 for oxygen storage. For

the Electrolysis Unit the SOEC technology is used in the baseline scenario. All production units

and energy storages are variables in the model and the ADAPT module in Sifre optimize the

production based on the economy of operation. This means the some production units can be

optimized out of the system, because they don’t generate enough value to pay for the invest-

ment cost. Sifre + ADAPT choses the most profitable sizes for all production units and for all

energy storages. The production is limited in two places. The dry wood inlet is limited to 500

MW and the Gas Turbine electricity production is limited to 200 MW.

The plant operation is optimized for one year (2030) using time steps at one hour.

9.1.2 Prices

The prices, described under the “Market Prices” section, is used in the baseline model. Sifre

calculate internally prices for the heat outputs from the plant. The prices are sometimes very

high compared to the fuel used. In calculation the total plant economy it has therefore been

chosen to operate with fixed prices for sale of District Heating, LT Process Heat and for the

District Heating Sink in all scenarios. This is done to be able to compare the plant operational

economy in different sensitivity scenarios. These prices are set to:

District Heating sale: 85 DKK/GJ

LT Process Heat sale: 120 DKK/GJ

District Heating Sink: 0 DKK/GJ

9.1.3 CAPEX

The total plant CAPEX is calculated by adding up the CAPEX of all the processes and adding a

30% contingency for plant cost not coupled directly to one production unit and for general

major uncertainties in the CAPEX data input. This is a very rough estimate like the CAPEX cost

for the processes, and it is strongly recommended to set up a more thorough study of all as-

pects of the plant cost.

NB: The ADAPT/SIfre simulation sets the size of the Water Shift Reactor Unit to 106 MW output

even though the unit has zero operation hours. It has not been possible to find the reason to

this in this project.

9.1.4 OPEX

The expenses and income related to product flows are calculated based on the Sifre energy

balance and the market prices set.

Fixed O&M are calculated from the size of the individual process steps and their specific fixed

O&M costs. Specific O&M cost are for some production units found in the Technology Data

Catalogues [6,7,8]. But for the main production units in the biomass-to-methanol process no

Dok. 17/01970-19

data are available in the Technology Data Catalogues. But for a Biomass-to-Methanol black box

process a 3% of CAPEX per year level is set in [8]. This value is used for all processes not listed

specific in the Technology Data Catalogues.

Variable O&M are found for the process steps available in the Technology Data Catalogues and

put in the Sifre model as taxes. For process steps not found in the Technology Data Catalogues

no variable O&M has been set. It is recommended to look more into this area.

9.1.5 Key economic figures

CAPEX expense per year is used to calculate the yearly revenue for the plant. The CAPEX ex-

pense is calculated by adding up the yearly payment based on lifetime and an interest at 4%

for each process step and adding the payment for the contingency over 30 years.

Yearly Revenue for the plant is calculated as the operational revenue (based on the energy

flows) subtracted the fixed and variable O&M and the yearly CAPEX expense. No taxes, credits,

grants or other posts are in-calculated at this level.

The IRR is estimated by finding the internal rate of return for a cash flow, where the invest-

ment (CAPEX) is split up over two years and the Yearly Revenue (without the CAPEX expense

per year) is put in as yearly income streams for 20 years after the two years of investment. This

is a very rough model, not taking reinvestment of equipment with shorter lifetime than 20

years or scrap values of equipment with longer lifetime into consideration. But as most equip-

ment groups have a lifetime at 20 years is assumed to be a fair simplification at this level.

Methanol Shadow Price is defined as the methanol selling price at which the Yearly Revenue is

zero.

Dok. 17/01970-19

9.1.6 Baseline model Energy Flows

Figure 46: Energy flows for baseline model in 20130

9.1.7 Baseline model Key Figures

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1121 1098 450

23 To Gas Boiler 0 0

215 215 To HT Proces Heat

300

947

1419 1457

0,0946 0,0946 1419 470

0 470

0,0

0

704 To DH

1418

From HTPH 280

0

0,0 67 Ngas 2999,5

256 0 LTPH 23

3962 3962 3000 1350 3430 3430

0 713

207 207 959 94 937 713

697,1

959 326 937

772,04 611

187 611

526

From EC

0,0 3465 0,0

256 To Water Shift 85

82,223

91

216 300,16

From ST 0,0 From CHP

502 142,7 From Water Shift 795

58 100,4

142,7 336

To DH

3465 450 331

35713571 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 47: Key figures for baseline model 2030

9.1.8 Baseline model Electrolysis Unit operation

Figure 48: Electrolysis Unit operation for baseline model in 2030

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.924

Thermal Gasification and GC 379 1.779 7.924

Methanol Synthesis and purification 433 571 7.924

Water Shift Reactor 106 57 -

Electrolysis Unit 170 750 8.323

Simple Cycle Gas Turbine 200 834 2.251

Steam Turbine 93 333 3.622

Gas Boiler 17 6 5.638

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 855 15

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.526

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.884

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.430 713 2.446

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 786 566 445

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 58 0 -

Wood Chips consumed 3.571 207 739

Electricity consumed 1.733 461 799

Natural Gas consumed 1.121 215 241

Other costs

O&M fixed 148

O&M var 85

CAPEX expense per year 397

Total 3.088 2.409

Yearly revenue 678

MeOH shadow price 515

IRR 16%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

9.2 Internal Rate of Return (IRR) at different MeOH prices

In this simulation the plant is optimized at different methanol price levels. The upper level is

the estimated future production cost of 2. Generation bioethanol at 828 DKK/MWh [14] and

the lower level is the estimated cost of gasoline plus CO2 at 598 DKK/MWh.

High MeOH price:

Figure 49: Energy flows for a high MeOH price scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1116 1092 448

23 To Gas Boiler 0 0

215 215 To HT Proces Heat

299

961

1435 1473

0,0957 0,0957 1435 470

0 470

0,0

0

841 To DH

1435

From HTPH 285

0

0,0 68 Ngas 3034,2

255 0 LTPH 23

4008 4008 3034 1366 3470 3470

0 828

207 207 970 95 948 828

705,1

970 330 948

780,96 618

189 618

532

From EC

0,0 4600 0,0

252 To Water Shift 86

66,96

77

168 298,59

From ST 0,0 From CHP

502 152,9 From Water Shift 814

58 100,4

152,9 343

To DH

4600 10584 216

36133613 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 50: Key figures for high MeOH price scenario

Figure 51: Electrolysis Unit operation at high MeOH price

Low MeOH price:

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 8.015

Thermal Gasification and GC 379 1.779 8.015

Methanol Synthesis and purification 433 571 8.015

Water Shift Reactor - - -

Electrolysis Unit 170 750 8.419

Simple Cycle Gas Turbine 200 834 2.239

Steam Turbine 91 326 3.780

Gas Boiler 18 7 5.269

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 847 14

Syn Gas Storage 0 0

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.462

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.801

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.470 828 2.873

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 791 564 446

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 58 0 -

Wood Chips consumed 3.613 207 748

Electricity consumed 1.752 466 817

Natural Gas consumed 1.116 215 240

Other costs

O&M fixed 145

O&M var 86

CAPEX expense per year 391

Total 3.516 2.427

Yearly revenue 1.089

MeOH shadow price 514

IRR 23%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

Figure 52: Energy flows for a low MeOH price scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1863 1435 589

427 To Gas Boiler 0 0

215 215 To HT Proces Heat

392

-739

0 0

0,0000 0 0 470

0 470

0,0

0

216 To DH

0

From HTPH 0

0

0,0 0 Ngas 0

0 0 LTPH 427

0 0 0 0 0 0

0 598

203 203 0 445 0 598

0

0 0 0

0,0845 0

0 0

0

From EC

0,0 295 0,0

638 To Water Shift 0

7,677

352

300 392,38

From ST 0,0 From CHP

502 150 From Water Shift 385

0 100,4

150,0 151

To DH

295 440 327

00 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 53: Key figures for low MeOH price scenario

Figure 54: Electrolysis Unit operation at low MeOH price

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 -

Thermal Gasification and GC - - -

Methanol Synthesis and purification - - -

Water Shift Reactor - - -

Electrolysis Unit - - -

Simple Cycle Gas Turbine 200 834 2.943

Steam Turbine 60 215 2.517

Gas Boiler 106 39 4.183

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 591 10

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 1.279

Total Plant CAPEX- incl. 30% cont. (MDKK) 1.662

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced - 598 -

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 739 576 426

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 0 0 -

Wood Chips consumed - 0 -

Electricity consumed - 0 -

Natural Gas consumed 1.863 215 400

Other costs

O&M fixed 36

O&M var 18

CAPEX expense per year 104

Total 623 558

Yearly revenue 64

MeOH shadow price -

IRR 8%

Energy efficiency (MeOH/input) 0%

Energy efficiency (MeOH+heat/input) 54%

Dok. 17/01970-19

Figure 55: Variations in MeOH selling price. Green dots are baseline scenario

9.3 Variation in electricity price and profile

9.3.1 More oscillating electricity prices

In the 2030 electricity price estimate over a year, the oscillations in the price are not that high.

With a much higher penetration of wind and PV, and a reluctance to invest in back-up capaci-

ties the oscillations could increase further. In the first sensitivity simulation, the average cost of

the electricity in the 2030 forecast is found and for each hour the price distance to the average

price is doubled. For negative prices, the price is set to 0 DKK/MWh as Sifre can’t operate with

negative prices. Figure 56 shows how the new price forecast looks like for the fourth of January

2030.

Figure 56: New electricity price forecast for 4.th Jan. 2030

Dok. 17/01970-19

9.3.1.1 Results

Figure 57: Energy flows for a 2x oscillating electricity price scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1451 1441 591

10 To Gas Boiler 0 0

215 215 To HT Proces Heat

394

760

1391 1442

0,0930 0,0928 1391 483

0 483

0,0

1,6895

695 To DH

1390

From HTPH 261

0,493

2,3 65 Ngas 2946,7

258 1 LTPH 10

3895 3895 2949 1327 3370 3370

0 713

207 207 942 79 921 713

685,3

943 320 921

759,1 600

184 600

517

From EC

0,0 3251 0,0

255 To Water Shift 84

123,52

117

290 393,79

From ST 0,6 From CHP

502 136,8 From Water Shift 852

73 100,4

136,8 363

To DH

3251 450 377

35113511 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 58: Key figures for 2x oscillating electricity price scenario

Figure 59: Electrolysis Unit operation at 2x oscillating electricity price scenario

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.790

Thermal Gasification and GC 379 1.779 7.790

Methanol Synthesis and purification 433 571 7.784

Water Shift Reactor 106 57 16

Electrolysis Unit 170 750 8.158

Simple Cycle Gas Turbine 200 834 2.953

Steam Turbine 93 333 3.908

Gas Boiler 11 4 7.166

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 1062 18

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.527

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.885

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.370 713 2.403

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 953 642 612

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 73 0 -

Wood Chips consumed 3.511 207 727

Electricity consumed 1.713 455 779

Natural Gas consumed 1.451 215 312

Other costs

O&M fixed 148

O&M var 88

CAPEX expense per year 397

Total 3.212 2.451

Yearly revenue 761

MeOH shadow price 487

IRR 17%

Energy efficiency (MeOH/input) 59%

Energy efficiency (MeOH+heat/input) 69%

Dok. 17/01970-19

Figure 60: Yearly revenue and MeOH shadow prices at double oscillating power prices

9.3.2 Lower electricity prices

In the second sensitivity calculation the electricity price in the baseline scenario is lowered by

100 DKK/MWh and 200 DKK/MWh. These simulations shows scenarios where the electricity

price still only reflects the marginal costs of power production and the VE penetration is very

high. Again the lower limit for the price is 0 due to Sifre.

9.3.2.1 Low electricity price

Figure 61: Energy flows for a baseline electricity price minus 100 DKK/MWh scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

161 161 66

0 To Gas Boiler 0 0

215 215 To HT Proces Heat

44

1520

1419 1475

0,0947 0,0947 1419 372

0 372

0,0

0

978 To DH

1419

From HTPH 264

0

0,0 3 Ngas 3000,4

266 0 LTPH 0

3963 3963 3000 1351 3431 3431

0 713

207 207 959 3 938 713

697,2

959 326 938

772,28 611

187 611

526

From EC

0,0 5663 0,0

213 To Water Shift 85

254,95

173

890 44,024

From ST 0,0 From CHP

502 55,16 From Water Shift 383

52 100,4

55,2 164

To DH

5663 442 578

35733573 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 62: Key figures for a baseline electricity price minus 100 DKK/MWh scenario

Figure 63: Electrolysis Unit operation at a baseline electricity price minus 100 DKK/MWh scenar-

io

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.926

Thermal Gasification and GC 379 1.779 7.926

Methanol Synthesis and purification 433 571 7.926

Water Shift Reactor - - -

Electrolysis Unit 170 750 8.325

Simple Cycle Gas Turbine 200 834 330

Steam Turbine 32 115 5.112

Gas Boiler 0 0 8.371

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 1074 18

Syn Gas Storage 0 0

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.249

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.523

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.431 713 2.447

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 230 465 107

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 52 0 -

Wood Chips consumed 3.573 207 740

Electricity consumed 1.751 361 632

Natural Gas consumed 161 215 35

Other costs

O&M fixed 138

O&M var 73

CAPEX expense per year 375

Total 2.751 1.993

Yearly revenue 758

MeOH shadow price 492

IRR 18%

Energy efficiency (MeOH/input) 65%

Energy efficiency (MeOH+heat/input) 77%

Dok. 17/01970-19

9.3.2.2 Very low electricity price

Figure 64: Energy flows for a baseline electricity price minus 200 DKK/MWh scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

71 71 29

0 To Gas Boiler 0 0

215 215 To HT Proces Heat

19

1579

1423 1486

0,0949 0,0949 1423 275

0 275

0,0

0

958 To DH

1423

From HTPH 257

0

0,0 0 Ngas 3008,6

257 0 LTPH 0

3974 3974 3009 1355 3441 3441

0 713

207 207 962 0 940 713

699,1

962 327 940

774,37 613

187 613

527

From EC

0,0 5578 0,0

166 To Water Shift 85

262,02

177

166 19,408

From ST 0,0 From CHP

502 44,96 From Water Shift 361

47 100,4

45,0 156

To DH

5578 442 187

35823582 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 65: Key figures for a baseline electricity price minus 200 DKK/MWh scenario

Figure 66: Electrolysis Unit operation at a baseline electricity price minus 200 DKK/MWh scenar-

io

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.948

Thermal Gasification and GC 379 1.779 7.948

Methanol Synthesis and purification 433 571 7.948

Water Shift Reactor - - -

Electrolysis Unit 170 750 8.348

Simple Cycle Gas Turbine 200 834 146

Steam Turbine 32 115 4.876

Gas Boiler - - -

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 1043 18

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.247

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.521

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.441 713 2.453

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 185 348 64

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 47 0 -

Wood Chips consumed 3.582 207 741

Electricity consumed 1.763 267 471

Natural Gas consumed 71 215 15

Other costs

O&M fixed 138

O&M var 72

CAPEX expense per year 375

Total 2.714 1.813

Yearly revenue 901

MeOH shadow price 451

IRR 21%

Energy efficiency (MeOH/input) 66%

Energy efficiency (MeOH+heat/input) 77%

Dok. 17/01970-19

Figure 67: Changes in MeOH and IRR at lower electricity price

9.4 Higher biomass cost

In this scenario the biomass cost is higher than expected due to high demand and limited pro-

duction. The biomass cost is raised by 20% and 40% compared to the baseline scenario.

9.4.1 Biomass price + 20% scenario

Figure 68: Energy flows for a high biomass price scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1119 1097 450

22 To Gas Boiler 0 0

215 215 To HT Proces Heat

300

941

1413 1451

0,0942 0,0942 1413 470

0 470

0,0

0

703 To DH

1412

From HTPH 279

0

0,0 67 Ngas 2986,8

292 0 LTPH 22

3945 3945 2987 1345 3416 3416

0 713

248 248 955 93 933 713

694,1

955 325 933

768,77 608

186 608

524

From EC

0,0 3040 0,0

255 To Water Shift 85

81,282

89

246 299,92

From ST 0,0 From CHP

502 143,9 From Water Shift 793

56 100,4

143,9 335

To DH

3040 450 398

35563556 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2

Storage

Dok. 17/01970-19

Figure 69: Key figures for high Biomass price scenario

Figure 70: Electrolysis Unit operation at high biomass price

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.890

Thermal Gasification and GC 379 1.779 7.890

Methanol Synthesis and purification 433 571 7.890

Water Shift Reactor 106 57 -

Electrolysis Unit 170 750 8.288

Simple Cycle Gas Turbine 200 834 2.249

Steam Turbine 93 333 3.612

Gas Boiler 17 6 5.547

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 855 15

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.526

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.884

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.416 713 2.435

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 785 565 444

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 56 0 -

Wood Chips consumed 3.556 248 883

Electricity consumed 1.726 460 793

Natural Gas consumed 1.119 215 240

Other costs

O&M fixed 148

O&M var 85

CAPEX expense per year 397

Total 3.076 2.547

Yearly revenue 530

MeOH shadow price 558

IRR 14%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

9.4.2 Biomass price + 40% scenario

Figure 71: Energy flows for a very high biomass price scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1863 1435 588

428 To Gas Boiler 0 0

215 215 To HT Proces Heat

392

-739

0 0

0,0000 0 0 470

0 470

0,0

0

215 To DH

0

From HTPH 0

0

0,0 0 Ngas 0

0 0 LTPH 428

0 0 0 0 0 0

0 713

285 285 0 445 0 713

0

0 0 0

0,0845 0

0 0

0

From EC

0,0 295 0,0

642 To Water Shift 0

6,7452

352

300 392,27

From ST 0,0 From CHP

502 150,3 From Water Shift 386

0 100,4

150,3 151

To DH

295 450 326

00 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 72: Key figures for very high Biomass price scenario

Figure 73: Electrolysis Unit operation at very high biomass price

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 -

Thermal Gasification and GC - - -

Methanol Synthesis and purification - - -

Water Shift Reactor - - -

Electrolysis Unit - - -

Simple Cycle Gas Turbine 200 834 2.942

Steam Turbine 60 215 2.523

Gas Boiler 106 39 4.189

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 591 10

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 1.279

Total Plant CAPEX- incl. 30% cont. (MDKK) 1.662

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced - 713 -

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 739 576 426

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 0 0 -

Wood Chips consumed - 0 -

Electricity consumed - 0 -

Natural Gas consumed 1.863 215 400

Other costs

O&M fixed 36

O&M var 18

CAPEX expense per year 104

Total 623 558

Yearly revenue 65

MeOH shadow price -

IRR 8%

Energy efficiency (MeOH/input) 0%

Energy efficiency (MeOH+heat/input) 54%

Dok. 17/01970-19

Figure 74: IRR and CAPEX at variation in biomass prices. Green dots is for the baseline scenario

9.5 Forced ASU operation – cost of flexibility

In the baseline model it is not feasible to invest in an oxygen production unit (ASU) and oxygen

storage. But such a unit and storage would increase the flexibility of the plant considerably, as

the biomass-to-methanol process would be able to operate without the Electrolysis Unit in

operation. Then the Electrolysis unit with up to 200 MW of electricity consumption would be

able to act directly in the regulation market without shutting down the total plant. The benefit

of such a capability is not taken into account in the Sifre/ADAPTS optimization. Therefore has

been chosen to calculate the cost of including this capability. This cost can then be compared

to estimated benefits of acting on the regulation market in 2030.

In this simulation a minimum size of the ASU at 0.013 MW has been set. This is same output of

oxygen as the Electrolysis Unit in the baseline scenario. Furthermore the Oxygen storage has

been set with a minimum capacity of 2.5 hours O2 production.

Dok. 17/01970-19

Figure 75: Energy flows for a forced ASU investment scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

14,4

0,000

1141 1100 451

41 To Gas Boiler 0,0074 0

215 215 To HT Proces Heat

301

940

1397 1434

0,1006 0,0932 1397 470

0 470

0,0

23,743

493 To DH

1397

From HTPH 278

6,93

32,4 0 Ngas 3156,5

371 9 LTPH 41

4212 4212 3189 1421 3610 3610

0 713

207 207 1019 42 986 713

741,1

1019 343 986

820,77 643

198 643

553

From EC

0,0 127 0,0

247 To Water Shift 90

101,71

63

230 300,66

From ST 8,6 From CHP

502 147,6 From Water Shift 825

52 100,4

147,6 349

To DH

127 30 246

37973797 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 76: Key figures a forced ASU investment scenario

Figure 77: Electrolysis Unit operation at a forced ASU investment price

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 8.424

Thermal Gasification and GC 379 1.779 8.424

Methanol Synthesis and purification 433 571 8.338

Water Shift Reactor 106 57 225

Electrolysis Unit 170 750 8.199

Simple Cycle Gas Turbine 200 834 2.255

Steam Turbine 93 333 3.761

Gas Boiler 17 6 2.543

Air Separation Unit 0,0114 182 647

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 855 15

Syn Gas Storage 0 0

Oxygen Storage 0,030 9

Total installed equip. CAPEX (MDKK) 4.717

Total Plant CAPEX- incl. 30% cont. (MDKK) 6.132

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.610 713 2.574

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 800 564 451

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 52 0 -

Wood Chips consumed 3.797 207 786

Electricity consumed 1.741 458 797

Natural Gas consumed 1.141 215 245

Other costs

O&M fixed 153

O&M var 90

CAPEX expense per year 414

Total 3.222 2.485

Yearly revenue 737

MeOH shadow price 509

IRR 17%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

Figure 78: Key Figures at forced ASU investment

The results for the forced ASU investment scenario are quite surprising. In no single modeled

scenario Sifre/ADAPT find a solution with ASU investment feasible. But in this scenario, where

the ASU investment is forced into the baseline model, the feasibility of the plant actually in-

creases despite of the higher investment cost.

It looks like the investment in extra flexibility is feasible without even taking the possible in-

come from acting on the regulation market into account.

9.6 Green Syn Fuel composition on syngas versus baseline composition

The H2/CO ratio of the raw syngas in the base case is: 20.1/15.8 = 1.27 [3]. This data is based

on a 25 bar gasification process. I the Green Sun Fuel report [2] the gasification pressure is

chosen to 10 bar. In this report, the H2/CO ratio is lower (28.6/26.7 = 1.07) [2]. The lower ratio

demands more H2 supply, and thereby more electrolysis operation. This is interesting because

it will change the H2/O2 consumption and result in an O2 surplus instead of a H2 surplus. An O2

surplus will open the operation of the electrolysis for more flexible operation due to electricity

market price changes.

A new H2/CO ratio from the gasifier demands a new SynGas/H2 ration to the Methanol Syn-

thesis.

The M-ratio still needs to be 2.05 at the Methanol Synthesis inlet

As described some hydrogen is recirculated from the purge gas. This recirculation raises the M-

ratio 0.13. So the M- ratio before the hydrogen is added has to be: 1.92.

As the CO2 content is 3% after the CO2 scrubber, the H2 content can be calculated:

1.92 = (x-3)/(97-x+3)

x = 66.8

Dok. 17/01970-19

The mole fraction of hydrogen has to be 66.8% at the inlet to the methanol synthesis and puri-

fication unit. To raise the level to 66.8% further 58.7 mole of H2 has to be added per 100 mole

raw syn gas.

This has to be converted to energy terms to be used in Sifre.

As CO represents approximately 46% of the syngas after CO2 and H2O removal, approximately

0.46 Mole of H2 has to be added to 1 Mole of cleaned syngas.

LHV:

CO: 283 kJ/Mole, H2: 244 kJ/Mole

1 Mole of cleaned syn gas has a LHV at: 255 kJ/Mole

0.587 Mole of H2 has a LHV at: 143 KJ

Thus the energy input to Methanol synthesis unit has to be: 64% cleaned syn gas and 36% H2.

By fixing this ratio in the Production Unit for Methanol Synthesis in Sifre, the simulation forces

to model to produce the needed H2 either by Electrolysis or by Water Shift Reaction.

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,016

1096 1070 439

27 To Gas Boiler 0 0

215 215 To HT Proces Heat

292

1303

1726 1772

0,0993 0,1151 1726 470

0 470

0,0

25,39

462 To DH

1726

From HTPH 343

7,411

34,6 0 Ngas 3113,2

337 9 LTPH 27

4158 4158 3148 1751 3836 3836

0 713

207 207 1006 28 1048 713

731,6

1006 365 1048

810,22 683

196 683

588

From EC

0,0 105 0,0

251 To Water Shift 95

72,601

61

159 292,39

From ST 9,2 From CHP

502 137,6 From Water Shift 775

62 100,4

137,6 328

To DH

105 0 214

37483748 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 79: Energy flows for a GSF SynGas composition scenario

Figure 80: Key figures for GSF SynGas composition scenario

Figure 81: Electrolysis Unit operation at GSF SynGas composition

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 8.316

Thermal Gasification and GC 379 1.779 8.316

Methanol Synthesis and purification 466 616 8.224

Water Shift Reactor 120 65 211

Electrolysis Unit 213 937 8.105

Simple Cycle Gas Turbine 200 417 2.193

Steam Turbine 91 326 3.614

Gas Boiler 12 4 2.351

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 843 14

Syn Gas Storage 0 0

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.339

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.640

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.836 713 2.735

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 767 565 433

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 62 0 -

Wood Chips consumed 3.748 207 776

Electricity consumed 2.069 457 946

Natural Gas consumed 1.096 215 236

Other costs

O&M fixed 154

O&M var 88

CAPEX expense per year 384

Total 3.366 2.584

Yearly revenue 782

MeOH shadow price 509

IRR 18%

Energy efficiency (MeOH/input) 62%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

Figure 82: Key figures at GSF SynGas composition

9.7 Investment cost variations

All equipment cost are in the first simulation reduced 30% and then in the second simulation

raised with 30%

9.7.1 Investment costs reduced with 30%

Figure 83: Energy flows for a low investment cost scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1121 1096 450

25 To Gas Boiler 0 0

215 215 To HT Proces Heat

300

944

1419 1457

0,0946 0,0946 1419 470

0 470

0,0

0

700 To DH

1419

From HTPH 281

0

0,0 68 Ngas 2999,7

255 0 LTPH 25

3962 3962 3000 1351 3431 3431

0 713

207 207 959 96 937 713

697,1

959 326 937

772,1 611

187 611

526

From EC

0,0 3439 0,0

253 To Water Shift 85

71,811

83

224 299,69

From ST 0,0 From CHP

502 150,7 From Water Shift 805

58 100,4

150,7 339

To DH

3439 450 367

35723572 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 84: Key figures for low investment cost scenario

Figure 85: Electrolysis Unit operation at low investment cost

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 126 7.924

Thermal Gasification and GC 379 1.245 7.924

Methanol Synthesis and purification 433 218 7.924

Water Shift Reactor 106 40 -

Electrolysis Unit 170 525 8.324

Simple Cycle Gas Turbine 200 584 2.248

Steam Turbine 93 233 3.657

Gas Boiler 18 5 5.373

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 999 12

Syn Gas Storage 0 -

Oxygen Storage 0,000 -

Total installed equip. CAPEX (MDKK) 2.988

Total Plant CAPEX- incl. 30% cont. (MDKK) 3.885

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.431 713 2.446

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 789 564 445

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 58 0 -

Wood Chips consumed 3.572 207 739

Electricity consumed 1.733 461 799

Natural Gas consumed 1.121 215 241

Other costs

O&M fixed 148

O&M var 85

CAPEX expense per year 262

Total 3.088 2.273

Yearly revenue 814

MeOH shadow price 476

IRR 24%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

9.7.2 Investment costs increased with 30%

Figure 86: Energy flows for a high investment cost scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1126 1105 453

20 To Gas Boiler 0 0

215 215 To HT Proces Heat

302

941

1419 1456

0,0946 0,0946 1419 470

0 470

0,0

0

701 To DH

1419

From HTPH 282

0

0,0 68 Ngas 2999,7

256 0 LTPH 20

3962 3962 3000 1351 3431 3431

0 713

207 207 959 92 937 713

697,1

959 326 937

772,1 611

187 611

526

From EC

0,0 3441 0,0

253 To Water Shift 85

77,124

84

561 302,14

From ST 0,0 From CHP

502 150,3 From Water Shift 801

59 100,4

150,3 338

To DH

3441 450 1056

35723572 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 87: Key figures for high investment cost scenario

Figure 88: Electrolysis Unit operation at high investment cost

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 234 7.924

Thermal Gasification and GC 379 2.313 7.924

Methanol Synthesis and purification 433 743 7.924

Water Shift Reactor 106 74 -

Electrolysis Unit 170 975 8.324

Simple Cycle Gas Turbine 200 1.085 2.266

Steam Turbine 93 433 3.639

Gas Boiler 16 7 5.913

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 759 17

Syn Gas Storage 0 -

Oxygen Storage 0,000 -

Total installed equip. CAPEX (MDKK) 5.881

Total Plant CAPEX- incl. 30% cont. (MDKK) 7.645

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.431 713 2.446

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 791 562 445

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 59 0 -

Wood Chips consumed 3.572 207 739

Electricity consumed 1.732 461 799

Natural Gas consumed 1.126 215 242

Other costs

O&M fixed 148

O&M var 85

CAPEX expense per year 516

Total 3.088 2.529

Yearly revenue 559

MeOH shadow price 550

IRR 12%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

Figure 89: Investment cost +/- 30%

9.8 Gas turbine investment halved

The CAPEX of the Gas Turbine unit is taken from the Technology Catalogue Data [7]. This cost is

for a standalone GT power plant including all necessary facilities. In this study a lot of these

common facilities is covered in the contingency on the overall CAPEX. It is therefore assumed,

that the CAPEX for the gas turbine is set to high in this study. This simulations shows the im-

pact of a 50% reduced GT CAPEX.

Figure 90: Energy flows for a low GT investment scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1119 1097 450

22 To Gas Boiler 0 0

215 215 To HT Proces Heat

300

945

1419 1457

0,0946 0,0946 1419 470

0 470

0,0

0

726 To DH

1419

From HTPH 281

0

0,0 68 Ngas 2999,7

255 0 LTPH 22

3962 3962 3000 1351 3431 3431

0 713

207 207 959 94 937 713

697,1

959 326 937

772,1 611

187 611

526

From EC

0,0 3439 0,0

252 To Water Shift 85

74,468

84

532 299,75

From ST 0,0 From CHP

502 150,7 From Water Shift 802

59 100,4

150,7 338

To DH

3439 450 900

35723572 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 91: Key figures for low GT investment cost scenario

Figure 92: Electrolysis Unit operation at low GT investment cost

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.924

Thermal Gasification and GC 379 1.779 7.924

Methanol Synthesis and purification 433 571 7.924

Water Shift Reactor 106 57 -

Electrolysis Unit 170 750 8.324

Simple Cycle Gas Turbine 200 417 2.248

Steam Turbine 93 333 3.642

Gas Boiler 17 6 5.637

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 855 15

Syn Gas Storage 0 0

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.109

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.341

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.431 713 2.446

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 788 562 443

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 59 0 -

Wood Chips consumed 3.572 207 739

Electricity consumed 1.733 461 799

Natural Gas consumed 1.119 215 240

Other costs

O&M fixed 148

O&M var 85

CAPEX expense per year 363

Total 3.086 2.375

Yearly revenue 711

MeOH shadow price 506

IRR 18%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

Figure 93: GT investment 50% down

9.9 SOEC versus Alkaline

In this simulation the SOEC electrolysis unit is replaced with an alkaline electrolysis unit. The

SOEC technology is under development and the cost estimation is very rough. The Alkaline

electrolysis is commercially available and the cost estimates therefore more precise. So a

switch to alkaline will reduce the technology risk of the total plant. The performance and eco-

nomic data for the alkaline electrolysis is found under the description of the Production Units.

Shifting from SOEC to Alkaline electrolysis changes the heat flow to the Electrolysis Unit. With

the SOEC technology HT process heat was supplied to the process while the alkaline technolo-

gy has a district heat output.

Figure 94: Energy flows for an alkaline electrolysis scenario

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

181 0 0

181 To Gas Boiler 0,0000 0

215 215 To HT Proces Heat

0

26

0 0

0,0676 0 0 470

0 470

0,0

598,1

555 To DH

0

From HTPH 0

174,6

815,6 0 Ngas 1328,4

270 218 LTPH 181

2832 2832 2144 598 1519 1519

0 713

206 206 685 188 415 713

498,3

685 144 415

551,86 271

133 271

233

From EC

0,0 224 0,0

261 To Water Shift 38

71,947

246

224 0

From ST 217,9 From CHP

500 113,3 From Water Shift 343

4 99,9

113,3 137

To DH

224 0 237

25532553 Wood-

Chips

Thermal

Gasification

SynGas

Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 95: Key figures for an alkaline electrolysis scenario

Shifting to alkaline electrolysis simply makes ADAPT/Sifre not invest in an electrolysis at the

given price.

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 5.664

Thermal Gasification and GC 379 1.779 5.664

Methanol Synthesis and purification 268 354 5.664

Water Shift Reactor 106 57 5.664

Electrolysis Unit - - -

Simple Cycle Gas Turbine - - -

Steam Turbine 33 118 4.167

Gas Boiler 41 15 4.609

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 720 12

Syn Gas Storage 0 0

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 2.516

Total Plant CAPEX- incl. 30% cont. (MDKK) 3.271

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 1.519 713 1.083

District Heat produced (fixed price 85 DKK/GJ) 500 306 153

Electricity produced 137 500 69

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 4 0 -

Wood Chips consumed 2.553 207 528

Electricity consumed 163 463 76

Natural Gas consumed 181 215 39

Other costs

O&M fixed 89

O&M var 51

CAPEX expense per year 226

Total 1.348 1.008

Yearly revenue 340

MeOH shadow price 489

IRR 15%

Energy efficiency (MeOH/input) 55%

Energy efficiency (MeOH+heat/input) 77%

Dok. 17/01970-19

10. References

[1] Martin Møller et al. Polygeneration, DONG Energy, PSO Report, 2011

[2] Danish Technological Institute, GreenSynFuel, final project report, EUDP 64011-0011, 2011

[3] Matteo Morandin et al. Methanol via biomass Gasification, Chalmers University og Tech-

nology, 2015

[4] Hannula and Kurkela, Liquid transportation fuels via large-scale fluidised-bed gasification of

lignocellulosic biomass, VTT, 2012

[5] Spath, Aden et al., Biomass to Hydrogen Production Detailed Design and Economics Utiliz-

ing the Battelle Columbus Laboratory Indirectly-Heated Gasifier, NREL 2005

[6] Anders Evald et al. Technology data for advanced bioenergy fuels, FORCE Technology, 2013

[7] Technology Data Catalogue for Energy Plants, aug. 2016

[8] Technology Data Catalogue for Energy Plants- UPDATE 2016

[9] Steven D. Phillips et al., Gasoline from Wood via Integrated Gasification, Synthesis, and

Methanol-to- Gasoline Technologies, NREL, 2011. [9]

[10] Yas A. Alsultanny et al. Oxygen Specific Power Consumption Comparison for Air Separation

Units, Engineering Journal, 2014

[11] Ryan M. Swanson et al., Techno-Economic Analysis of Biofuels Production Based on Gasifi-

cation, NREL, 2010

[12] Jonathan Chou et. al. Compressed Air Energy Storage, University of Pennsylvania, 2014

[13] Analyseforudsætninger, Energinet, 2016

[14] EA Energianalyse, Samfundsøkonomiske aspekter ved produktion af 2. generations

bioethanol, 2009

[15] GRØN ROADMAP 2030. SCENARIER OG VIRKEMIDLER TIL OMSTILLING AF

TRANSPORTSEKTORENS ENERGIFORBRUG, November 2015:

[G] Internal estimated value.

Dok. 17/01970-19

11. Appendix 1

During the sensitivity simulations the ADAPT/Sifre tool was updated. All simulations up to and

including: Forced ASU operation – cost of flexibility was performed with the Sifre Frontend

300.1.1 version and all simulations descriped after Forced ASU operation – cost of flexibility has

been performed with the Sifre Frontend 300.2.0 version.

The simulation of the Energy Plant Type III baseline model in the old version and in the new

version is compared hereunder. The output data shows minor differences but not really chang-

ing the optimization results.

11.1.1 Baseline model Energy Flows

Figure 96: Energy flows for baseline model in 2030

11.1.2 Baseline model Key Figures

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1121 1098 450

23 To Gas Boiler 0 0

215 215 To HT Proces Heat

300

947

1419 1457

0,0946 0,0946 1419 470

0 470

0,0

0

704 To DH

1418

From HTPH 280

0

0,0 67 Ngas 2999,5

256 0 LTPH 23

3962 3962 3000 1350 3430 3430

0 713

207 207 959 94 937 713

697,1

959 326 937

772,04 611

187 611

526

From EC

0,0 3465 0,0

256 To Water Shift 85

82,223

91

216 300,16

From ST 0,0 From CHP

502 142,7 From Water Shift 795

58 100,4

142,7 336

To DH

3465 450 331

35713571 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 97: Key figures for baseline model 2030

11.1.3 Baseline model Electrolysis Unit operation

Figure 98: Electrolysis Unit operation for baseline model in 2030

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.924

Thermal Gasification and GC 379 1.779 7.924

Methanol Synthesis and purification 433 571 7.924

Water Shift Reactor 106 57 -

Electrolysis Unit 170 750 8.323

Simple Cycle Gas Turbine 200 834 2.251

Steam Turbine 93 333 3.622

Gas Boiler 17 6 5.638

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 855 15

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.526

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.884

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.430 713 2.446

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 786 566 445

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 58 0 -

Wood Chips consumed 3.571 207 739

Electricity consumed 1.733 461 799

Natural Gas consumed 1.121 215 241

Other costs

O&M fixed 148

O&M var 85

CAPEX expense per year 397

Total 3.088 2.409

Yearly revenue 678

MeOH shadow price 515

IRR 16%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

11.1.4 Baseline model Energy Flows new Sifre version

Figure 99: Energy flows for baseline model in 2030 new Sifre version

11.1.5

11.1.6 Baseline model Key Figures new Sifre version

Black numbers are Energy flows in GWh for 2030 0

Red numbers are prices in DKK/MWh

0,0

0,000

1106 1085 445

22 To Gas Boiler 0 0

215 215 To HT Proces Heat

296

950

1418 1456

0,0946 0,0946 1418 470

0 470

0,0

0

703 To DH

1418

From HTPH 281

0

0,0 67 Ngas 2998,5

256 0 LTPH 22

3961 3961 2998 1350 3429 3429

0 713

207 207 958 93 937 713

696,9

958 326 937

771,78 611

187 611

526

From EC

0,0 3451 0,0

253 To Water Shift 85

73,634

81

210 296,45

From ST 0,0 From CHP

502 151,3 From Water Shift 799

57 100,4

151,3 337

To DH

3451 450 323

35703570 Wood-

Chips

Thermal

Gasification

SynGas Storage

EC EL

SynGas

LT Proces

Heat

Proces-

heat

El

DK1

Gas-

Boiler

GT SC

Methanol

Synthesis

District

HeatingDH

storage

DH Sink

Methanol

Market

Methan

ol

O2 EC gasEC gas

splitter

H2

O2

Market

Water shift

Dryer

DH

Demand

NGas

Dried

Wood

Wood-

Chips

Market

NGas

market

HT Proces

Heat

HM1

H Splitter 1

HM2

H Splitter 2 HM3

H Splitter 3

Steam

Turbine

ASU

O2Storage

Dok. 17/01970-19

Figure 100: Key figures for baseline model 2030 new Sifre version

11.1.7 Baseline model Electrolysis Unit operation new Sifre version

Figure 101: Electrolysis Unit operation for baseline model in 2030 new Sifre version

Plant SummaryProduction units Opt Size Plant cost FLE Op h

MW p Out. MDKK h/y

Wood Chip Dryer 500 180 7.921

Thermal Gasification and GC 379 1.779 7.921

Methanol Synthesis and purification 433 571 7.921

Water Shift Reactor 106 57 -

Electrolysis Unit 170 750 8.320

Simple Cycle Gas Turbine 200 834 2.223

Steam Turbine 93 333 3.628

Gas Boiler 17 6 5.548

Air Separation Unit - - -

Energy Storages Opt. Size Plant cost

MWh MDKK

District Heat Storage 855 15

Syn Gas Storage 0 -

Oxygen Storage 0 -

Total installed equip. CAPEX (MDKK) 4.526

Total Plant CAPEX- incl. 30% cont. (MDKK) 5.884

Production and consumption Quantity Price Income Expenses

GWh/y DKK/MWh MDKK/y MDKK/y

Methanol produced 3.429 713 2.445

District Heat produced (fixed price 85 DKK/GJ) 502 306 154

Electricity produced 781 563 440

Process Heat produced (fixed price 120 DKK/GJ) 100 432 43

Heat sink (fixed price 0 DKK/GJ) 57 0 -

Wood Chips consumed 3.570 207 739

Electricity consumed 1.732 461 798

Natural Gas consumed 1.106 215 238

Other costs

O&M fixed 148

O&M var 85

CAPEX expense per year 397

Total 3.082 2.404

Yearly revenue 678

MeOH shadow price 515

IRR 16%

Energy efficiency (MeOH/input) 61%

Energy efficiency (MeOH+heat/input) 72%

Dok. 17/01970-19

Dok. 17/01970-19